Many computer systems, both static and portable, have been designed so that a user may enter data by means of a pen and a digitizer. Software exists to translate handwriting into recognized standard text. Many applications exist that are capable of taking advantage of pen input. Pen input facilitates the use of computers by those unfamiliar with or unskilled in the use of computer keyboards. Moreover, the use of pen-based computers, and the storage and transport of information in digital form, realizes an important commercial benefit--the reduction or elimination of the use of paper.
Digitizers typically sample the position of the pen tip around one hundred times a second, and are sensitive to movements of one seventieth of an inch. They are thus capable of very accurately recording the movement of the human hand. Computer signature verification can exploit this by analyzing not only the visible shape of the signature but also dynamic aspects such as speed and rhythm.
Algorithms exist that can take pen-based input (such as a handwritten signature), determine the fundamental characteristics of the pen-based input, and represent the characteristics of the pen-based input in an electronic format. Algorithms also exist that can determine if handwritten signatures in electronic format are that of the same person. For example, see U.S. Pat. No. 5,109,426 (U.K. Application No. 90 24383.3), U.S. Pat. No. 4,495,644 and U.K. Application No. 1480066, all expressly incorporated by reference herein.
Signature verification can make a highly significant contribution to computer security, in that all other security mechanisms rely upon what a person knows (e.g., a password) or possesses (e.g., a physical key). By relying instead on an aspect of physical behavior which cannot be stolen or divulged, signature verification offers secure evidence as to the real identity of the user.
To date, signature verification has been employed mainly in the area of access security, with the object of verifying the identity of an individual before giving the user access to all or part of a computer system.
However, traditional signatures made on a piece of paper are used to witness intentions in such contexts as signing a contract or will, and as a shield against repudiation as when signing a money order.
There are many areas today where, despite the availability of computerized documents, it is necessary to rely upon paper because of the legal or cultural requirement for a signature.
Thus, it is often the case that a hardcopy of a document is preferred to that same document in a digital or electronic format. For example, a will or contract for the transfer of land is required by law in most jurisdictions to be in writing and to include original handwritten signatures of the parties and witnesses to the document. When a document is in electronic form, because it is relatively easy to manipulate the contents of the document, it is often uncertain if a document viewed at a later date is the same as the document originally created. Although handwritten signatures captured using pen input facilities can be incorporated in the text of such documents, one is never certain if a document viewed at a later date is the one that was "electronically" signed.
Accordingly, it would be desirable to apply the science of handwritten signature capture and verification to a much wider context than as a security access mechanism. In particular, there exists a need in the area of testifying to an intention (such as, for example, signing a legal document) for a secure signature capture and verification method that relates the document signed to the signature of the signer.
Existing systems have focused on whether an electronic version of a signature has been manipulated after it was created and whether an electronic version of a signature associated with an electronic document was captured at the time of a transaction to which it relates. For example, U.S. Pat. No. 5,195,133 to Kapp et al. describes a mechanism that attempts to assure that a signature purportedly approving a commercial obligation was captured at the time of a questioned transaction and is not a genuine signature obtained on some other occasion and fraudulently merged into the digital record of the transaction. The apparatus of the Kapp et al. patent creates a digital record of a transaction, captures a digital representation of a signature at the time of the transaction, and then uses the digital record of the transaction to encrypt the digital representation of the signature. This method aims to ensure that the representation of the signature was made when it is said it was made. However, such a system does not verify if the document that was signed using a digitally captured handwritten signature has been later modified. Moreover, systems such as the Kapp et al. system require a transaction, and are incapable of operation where a signature is to be captured and verified in an environment unrelated to a transaction.
Existing handwritten capture and verification systems are designed for use on a single platform. Often, the handwritten signature is encoded in such a way that other applications are not capable of utilizing the electronic form of the handwritten signature. By virtue of today's advanced computer-to-computer communications, including communication over the Internet, many applications will not require that verification be performed upon the same machine or at the same time as the act of signing itself. For example, it would be desirable for a system to enable a handwritten signature to be captured electronically on one device, stored, electronically transmitted to another device on another computer platform, and later verified. Accordingly, there is a need for an integrated cross-platform signature verification system. In particular, there is a need for a system that does not presuppose any particular underlying hardware, and is designed to be portable across different types of computer and operating system.
Many businesses and government departments often require people to sign documents. For example, when buying goods by check or credit card, when signing a car rental agreement, when entering a lease, when applying for a driver's license or other government permit, on election day, or to certify attendance at an examination. Often, the person requesting the signature does not know the individual who is required to sign, and does not have an authentic signature of the signer to compare with the requested signature. Moreover, even if an authentic signature is available for comparison, the person requesting the signature often is unskilled in determining whether two signatures are from the same person. Accordingly, there is a need for a system that allows a signature to be captured electronically in one location, electronically transmitted to a central location that has recorded verified signing behaviors of many individuals, and returns an indication of the identity of the signer.
In certain situations, a person who has signed a contract or other legal document will attempt to terminate his or her obligations by claiming at a later date that he or she did not understand the nature of the document being signed or that he or she was misled when signing the document. Moreover, in a multi-windowed computing environment, a person signing a document electronically may not be sure which document stored on the computer he or she is actually signing. It would be useful if a record was made at the time of signing (that could later be retrieved) that records what the signer was told when signing a document and, before signature, alerts the signer as to the identity, nature and gravity of the document being signed.
In short, there is a need for a system that takes advantage of the increasing availability of these penbased input devices by enabling the application of handwritten signature capture and verification technology to be used in the diverse contexts where signature capture is needed.